Skip to content

Advancing FAIR and GO FAIR in the U.S.: Notes from the Workshop

Posted in conf, and events

GO FAIR Professional Development Workshop to Advance FAIR and GO FAIR in the US, held at Georgia Tech’s Coda building, Atlanta, GA, February 25-27, 2020

In February, I had the privilege of attending “Advancing FAIR and GO FAIR in the U.S.”, a workshop put on by GO FAIR USA in conjunction with the San Diego Supercomputer Center’s (SDSC) Research Data Services, and funded by the National Science Foundation (NSF). The event was held in Atlanta, and hosted by the South Big Data Hub. Twenty-eight research data professionals from academia, government, industry, and nonprofits gathered to learn from the experiences of European GO FAIR founders, Albert Mons (Phortos Consultants), and Luiz Bonino and Erik Schultes (GO FAIR International Support and Coordination Office (GFISCO)).  

Little did we know at the time, but the workshop’s significance was heightened due to the fact that this was the last time many of us were able to gather as a community for the foreseeable future given the rapidly developing COVID-19 pandemic. In parallel with the workshop, a new GO FAIR implementation network has been founded to help with research surrounding COVID-19: Virus Outbreak Data Network (VODAN), of which GO FAIR US is one of the founding members. A webinar series is also currently underway to explore the work of VODAN. 

The goals of the workshop were to: 

  • Facilitate development of a community of practice for FAIR awareness and capacity-building in the US;
  • Improve understanding of technologies enabling FAIR data, and how to teach them to others; and,
  • Preparation for teaching or supporting FAIR data management and policies for researchers, local institutions, professional organizations, and others.

More information regarding the workshop and the US GO FAIR office can be found on their website.

Below are four key themes I took away from the workshop. 

1. The expanding role of the Data Steward: In her introduction to the workshop, Melissa Cragin (from the San Diego Supercomputer Center) noted that those working in research data management in Europe are generally less focused on the task of data curation, but rather emphasize the centrality of data stewardship and the FAIRification of data in the work of FAIR data professionals. Albert Mons continued the theme with a discussion of the FAIR Data Stewardship model, which seeks to achieve a unified organizational outlook with consistent names and definitions of what FAIR data is and how to achieve it. 

For those interested in reading more about the role of the data steward, Barend Mons lays out the importance of the data steward and advocates the allocation of 5% of research funds to cover the costs of FAIRifing data in this recent Nature opinion piece. Additionally, I found the 2019 report by the LCRDM Data Stewardship Task Group particularly insightful in understanding the role of the data steward as practiced in Dutch institutions and how this supports the implementation of FAIR data policies and procedures for the FAIRification of research data. 

2. Machine actionable data is key. Building on the importance of the Data Steward in the GO FAIR model, throughout the workshop there was particular emphasis on the FAIRification of data. As Albert Mons put it, “FAIR data is data that can be used and understood efficiently by people but mostly by machines.” Luiz Bonino hit a similar tune as he stated: “The data needs to be understandable by machines, and the machines need to be built to understand the data.” 

When speaking about the technical fundamentals of FAIR, Bonino acknowledged that it is currently easier to focus on the FAIR principles in terms of metadata rather than data at this point, noting that semantic interoperability is the biggest challenge to achieving the ultimate goal of interoperable, FAIR, data. The requirements of FAIRifing data was acknowledged as being decidedly discipline specific, although work has been done to create a general workflow to describe the steps involved and is outlined in a recent article, “A Generic Workflow for the Data FAIRification Process.”’

3. “There is no such thing as data which is 100% FAIR”: In order to measure the FAIRness of data, the GO FAIR initiative is developing a set of evaluators to enable machines to assess a given dataset and to determine to what degree it is FAIR.  As Bonino noted in his presentation, since FAIR is centered around machine actionability, it should follow that machines assess the FAIRness of those data. The tool currently includes the ability to design your own metric tests, thus allowing communities to make domain-specific indicators. The tool is available here and the recent RDA Working Group analysis of the indicator tool final report is available here.   

4. How best to apply FAIR principles in the US?: Offering the NSF perspective on FAIR, NSF Science Advisor for Public Access Beth Plale stated that the FAIR principles are now part of the standard vocabularies of federal agencies and have made a significant impact on how agencies talk about open access. She also noted that although it is relatively early now in the US, we are drawing on existing European practices and that uptake is growing over time. Christine Kirkpatrick (from GO FAIR US and the San Diego Supercomputer Center) echoed this sentiment, stating that she is seeing general enthusiasm for FAIR in the US, but thus far, serious implementation and the funding to support such work has yet to materialize. 

Whether FAIR implementations in the US take the shape of the European GO FAIR model and focus on data stewardship is to be determined; however, the maturing practice in the US makes the time ripe for a concerted effort to institutionalize FAIR data practices via advocacy, education and infrastructure development. 

Since the workshop, participants have continued the conversation via a slack channel. Current work amongst the group has centered on adapting existing materials and identifying new curriculum needs (perhaps revamping the Library Carpentry: FAIR Data and Software course),  and developing a new US GO FAIR website

The workshop was packed with information and thoughtful discussion, and much more than summarized above was covered. I hope that other participants will publish their thoughts and perspectives on the workshop and FAIR in the US in the coming weeks. To get involved or to learn more go to or request to be added to the #gofairUS Slack channel.

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *